Abstract
Introduction
Minimally invasive ventral hernia repair (MISVHR) has been performed for almost 30 years; recently, there has been an accelerated adoption of the robotic platform leading to renewed comparisons to open ventral hernia repair (OVHR). The present study evaluates patterns and outcomes of readmissions for MISVHR and OVHR patients.
Methods
The Nationwide Readmissions Database (NRD) was queried for patients undergoing OVHR and MISVHR from 2016 to 2018. Demographic characteristics, complications, and 90-day readmissions were determined. A subgroup analysis was performed to compare robotic ventral hernia repair (RVHR) vs laparoscopic hernia repair (LVHR). Standard statistical methods and logistic regression were used.
Results
Over the 3-year period, there were 25 795 MISVHR and 180 635 OVHR admissions. Minimally invasive ventral hernia repair was associated with a lower rate of 90-day readmission (11.3% vs 17.3%, P < .01), length of stay (LOS) (4.0 vs 7.9 days, P < .01), and hospital charges ($68,240 ± 75 680 vs $87,701 ± 73 165, P < .01), which remained true when elective and non-elective repairs were evaluated independently. Postoperative infection was the most common reason for readmission but was less common in the MISVHR group (8.4% vs 16.8%, P < .01). Robotic ventral hernia repair increased over the 3-year period and was associated with decreased LOS (3.7 vs 4.1 days, P < .01) and comparable readmissions (11.3% vs 11.2%, P = .74) to LVHR, but was nearly $20,000 more expensive. In logistic regression, OVHR, non-elective operation, urban-teaching hospital, increased LOS, comorbidities, and payer type were predictive of readmission.
Conclusions
Open ventral hernia repair was associated with increased LOS and increased readmissions compared to MISVHR. Robotic ventral hernia repair had comparable readmissions and decreased LOS to LVHR, but it was more expensive.
Introduction
Abdominal wall hernias occur at a high rate and require significant surgical resources to treat. In the United States, over 2 million laparotomies are performed each year, and as many as 1 in 4 laparotomies result in an incisional hernia. 1 An estimated 500 000 ventral hernias are repaired annually in this country making it one of the most common surgeries performed by general surgeons. 2 Ventral hernia repair (VHR) carries the risk of wound complications such as surgical site infection (SSI) and other surgery and hospital-acquired problems. 3 Surgical site infection during a VHR increases the risk of recurrence, which subsequently increases the risk of complications and recurrence in any subsequent VHR. 4 This can create a “vicious cycle” of complications and recurrence that perpetuates morbidity and cost. With an annual cost to the US health care system of over $3.2 billion, a 1% reduction in VHR operations could save as much as $32 million. 2
Minimally invasive approaches to VHR (MISVHR), which include both laparoscopic and robotic surgery, offer a risk of recurrence comparable to open techniques but with decreased rates of SSI and potentially shorter hospital length of stay (LOS). 5 Laparoscopic ventral hernia repair (LVHR) has been limited mostly to patients who have smaller defect sizes or by surgeon technical ability. 6 Robotic ventral hernia repair (RVHR) appears to have improved ergonomics and suturing within the abdomen, which specifically impacts the ability to close the abdominal wall and carries a similar perioperative risk profile as compared to laparoscopy. These factors, specific to the robotic approach, may allow for surgeons to operate on patients with larger hernia defects than traditional LVHR and consistently achieve fascial closure. 7 There is, however, an increased cost of the robotic platform without clearly demonstrated clinical benefit. 8 Significant variability in the application of minimally invasive surgery (MIS) techniques exists due to differences in surgeon training and institutional resources. MIS may be particularly advantageous to high-risk patients with amenable hernia defects; however, the ideal patient population for MIS vs open techniques has yet to be defined.
Further studies of these evolving techniques are required to improve patient selection and ultimately surgical outcomes. The purpose of this study is to compare nationwide patterns and outcomes of readmission for patients who underwent open ventral hernia repair (OVHR) vs MISVHR. The authors hypothesized that patients who undergo MISVHR are at lower risk for short-term readmission in comparison to patients who undergo OVHR. Furthermore, the authors hypothesized that there will not be any difference in readmission rates when specifically comparing LVHR and RVHR.
Methods
Study Population
Institutional Review Board approval was obtained prior to the start of the study. The Nationwide Readmissions Database (NRD) was used to identify patients who underwent ventral hernia repair over a recent 3-year period (2016-2018). The NRD is a publicly available database maintained by the Healthcare Cost Utilization Project (HCUP) that allows for nationwide tracking of patients at their index hospitalization and during subsequent readmission(s). 9 Unweighted, the NRD consists of discharge data from approximately eighteen million patients each year. Only patients who are admitted to the hospital are included; patients who undergo outpatient surgeries are not part of the NRD.
Patients were retrospectively identified in the database using International Classification of Disease 10th Revision Procedure Coding System (ICD-10-PCS) codes. For OVHR, the ICD-10-PCS codes 0WQF4ZZ, 0WUF07Z, 0WUF0JZ, and 0WUF0KZ were used, and for LVHR the ICD-10-PCS codes 0WUF47Z, 0WUF4JZ, 0WUF4KZ were used. A subset of patients who underwent robotic repair were identified using abdominal robotic modifier codes—8E0W0CZ, 8E0W3CZ, 8E0W4CZ, 8E0W7CZ, 8E0W8CZ, and 8E0WXCZ. Patients were excluded from analysis if they received an operation in the last 3 months of the calendar year. For patients who had their operations in October-December, the NRD could not identify hospital readmissions in the following calendar year. For the readmission analysis, patients were excluded if they suffered a mortality at their index hospitalization.
Outcomes
The primary outcome that was assessed in the study was 90-day readmission rate, which was compared between OVHR and MISVHR patients. Readmission rates were further compared between elective and non-elective repairs. Secondary outcomes included hospital LOS, diagnoses associated with readmission, and procedures performed once readmitted. Diagnoses and procedures associated with readmission were identified using ICD-10-Clinical Modification and ICD-10-PCS codes, respectively. The authors evaluated the top 5 ICD-10-CM codes as the reasons for readmission for each patient and the top ten ICD-10-PCS codes for procedures performed once readmitted. Other demographic information was tracked including age, comorbidities, in-hospital mortality, teaching hospital status, payer type, and income quartile. A weighted Charlson Comorbidity Index was calculated using the method of Quan to determine risk of mortality. 10 A subgroup analysis was performed to compare 90-day readmissions for patients undergoing LVHR with those undergoing RVHR.
Statistical Analyses
All statistical analyses were completed by a trained statistician. Categorical variables were reported as percentages, and continuous variables were reported as means with corresponding standard deviations. A univariate comparison was performed between patients who underwent OVHR and those who underwent MISVHR. A subgroup univariate analysis was then done to compare LVHR and RVHR. Chi-square tests were used to evaluate categorical variables and continuous variables were evaluated using an analysis of variance test. Logistic regression was then used to determine factors that were predictive of readmission. Logistic regression accounted for all of the main output variables in the database and those previously linked to readmission, including age, type of hernia repair, elective case status, LOS, comorbidities, hospital type, and payer type. 11 Statistical significance was set to P < .05, which was two-sided.
Results
Demographics and Hospital Characteristics for MISVHR vs OVHR Patients.
Significance was set to P < 0.05.Abbreviations: MISVHR = minimally invasive ventral hernia repair, OVHR = open ventral hernia repair, CST = component separation technique, CCI = charlson comorbidity index.
The 90-day readmission rate for MISVHR was decreased in comparison to OVHR (11.3% vs 17.3%, P < .01) (Figure 1). The most common reasons for readmissions were similar for both MISVHR and OVHR patients and can be found in Table 2. Postoperative infection was the most common reason for readmission and was nearly doubled in the OVHR cohort (8.4% vs 16.8%, P < .01). The next 3 most common reasons for readmission in each group were kidney failure, sepsis, and intestinal obstruction. Unlike for OVHR, removal of mesh (3.7%) was the most common procedure performed on readmitted MISVHR patients (Table 3). The most common procedure performed on readmitted OVHR patients was an open abdominal wall debridement (4.6%). The total readmission charges were similar between MISVHR and OVHR patients ($62,590 ± 130 481 vs $64,357 ± 107 328, P = .40). The rate of 90-day readmissions was significantly decreased among patients undergoing. Minimally invasive ventral hernia repair compared to open repair; however, there was no difference between laparoscopic and robotic repair. MIS = minimally invasive surgery. Most Common Diagnoses for Readmitted Patients. Significance was set to P < 0.05. Most Common Procedures for Readmitted Patients. Significance was set to P < 0.05.
Minimally Invasive Ventral Hernia Repair (MISVHR) vs OVHR Patients: Elective vs Non-Elective Repair.
Significance was set to P < 0.05.Abbreviations: MISVHR = minimally invasive ventral hernia repair, OVHR = open ventral hernia repair, CCI = charlson comorbidity index.
Robotic vs Laparoscopic Repair
Laparoscopic Ventral Hernia Repair (LVHR) vs RVHR Subgroup Analysis: Overall Comparison.
Significance was set to P < 0.05.Abbreviations: LVHR = laparoscopic ventral hernia repair, RVHR = robotic ventral hernia repair, CCI = charlson comorbidity index.
Laparoscopic Ventral Hernia Repair (LVHR) vs RVHR Subgroup Analysis: Elective and Non-Elective Cases.
Significance was set to P < 0.05.Abbreviations: LVHR = laparoscopic ventral hernia repair, RVHR = robotic ventral hernia repair, CCI = charlson comorbidity index.
Logistic Regression
In logistic regression, after controlling for confounding variables, OVHR was independently associated with a higher rate of 90-day readmission than MISVHR (OR 1.43, 95% CI 1.37-1.49). However, there were several other factors that positively correlated with short-term readmission. Factors that were associated with readmission included diabetes (OR 1.14, 95% CI 1.12-1.17), smoking (OR 1.08, 95% CI 1.05-1.12), older age (OR 1.01, 1.01-1.02), LOS (OR 1.02, 95% CI 1.02-1.03), urban-teaching hospital status (OR 1.23, 1.17-1.30), non-elective surgery (OR 1.09, 95% CI 1.07-1.12), and Medicare (OR 1.34, 95% CI 1.24-1.45) or Medicaid (OR 1.25, 95% CI 1.16-1.36) payer types. Factors that did not correlate with readmission in logistic regression included obesity, day of week that the operation was performed, and self-pay status (each P > .05).
Discussion
Ventral hernia repair is one of the most common operations performed by general surgeons nationwide and accounts for considerable cost to the health care system. 12 Postoperative complications can result in frequent and prolonged utilization of health care resources and subsequent failure of the procedure. The goals of any VHR should be to prevent iatrogenic and other complications, achieve fascial closure, and ultimately improve quality of life (QOL). Unfortunately, a considerable number of patients have no change in QOL evaluations despite hernia repair. 13 Given that readmission rates for surgeons in some areas have become a marker of quality care and even used for a financial incentive or penalty, evaluating short-term readmission rates per procedure type is important and allows for a relative determination of the short-term efficacy of the repair and assessment of complication rates. In the present study, patients undergoing MISVHR had a decreased rate of 90-day readmissions and decreased LOS when compared to OVHR patients; this was true for both elective and non-elective repair. Minimally invasive ventral hernia repair patients had less postoperative infection than patients undergoing OVHR despite having higher rates of diabetes and obesity. When LVHR and RVHR were compared in a subgroup analysis, readmission rates were comparable; however, LVHR was significantly cheaper to perform. Notably, both LVHR and RVHR had lower charges than OVHR, with less difference noted when limited to the subset of elective cases in Table 4. Increased LOS likely plays a role in the difference between MISVHR and OVHR.
In previous large-scale studies, MISVHR has demonstrated decreased risk of postoperative wound infection and LOS in comparison to OVHR. For instance, Colavita et al used data from the International Hernia Mesh Registry to compare short-term and long-term outcomes for patients undergoing OVHR vs LVHR. 14 The authors found an overall increased SSI rate for OVHR vs LVHR, and LOS that was 2 days shorter in the laparoscopic group. Additionally, despite having increased pain in the short term (perhaps due to laparoscopic tacking, full thickness sutures, or the lack of fascial closure), quality QOL was comparable in long term between the 2 groups when measured with the Carolinas Comfort Scale.15,16 A later meta-analysis of randomized controlled trials by Sajid et al concluded that LVHR resulted in fewer overall postoperative complications and potentially even decreased operative time. 17 The findings in this paper support the previous findings and demonstrate that readmissions are also decreased in LVHR patients, which are predominantly driven by infection.
Despite the aforementioned benefits of MIS repair, OVHR will continue to have a prominent place in herniorrhaphy. In this study, OVHR was performed on patients who were more comorbid on average. There are factors that were not captured by the NRD that may influence a surgeon’s choice about whether to perform an open repair. For patients who have large and complex hernias (eg, patients with infection, mesh fistula, substantial loss of domain, multiply recurrent or incarcerated hernias, and strangulated intestine), OVHR likely remains the most common route of repair for most general surgeons. There are also indications for OVHR beyond complexity, including the subset of patients who cannot tolerate laparoscopic insufflation. Open ventral hernia repair can be performed concomitantly with other procedures, such as panniculectomy. 18 Even for the most talented MIS surgeons, a firm understanding and dexterity with open hernia repair remains necessary. The challenge for the surgeon is to choose the repair that will be best for the patient and is within the surgeon’s skill set.
Robotic ventral hernia repair is one of the most rapidly evolving fields in abdominal wall reconstruction, and the robotic approach has allowed for surgeons to employ MIS surgical techniques on more complex patients. Traditionally, LVHR has involved hernias that are small to medium sized hernias and relied on intraperitoneal placement of a coated synthetic mesh, with or without fascial closure. Robotic ventral hernia repair has allowed for an MIS approach to larger hernia types with the potential to place mesh in the extra-peritoneal position (eg, robotic transabdominal preperitoneal repair, robotic retrorectus repair, etc) or even perform component separation. 19 In the present study, over one-fifth of MIS patients received RVHR from 2016 to 2018 and increased over the course of the study. Length of stay appears to be one of the main benefits in comparison to OVHR, but the fact that RVHR had a shorter LOS compared to LVHR is somewhat surprising and cannot be directly explained by available data. In a randomized study comparing LVHR vs RVHR, patients were found to have comparable short-term quality of life, complications, and readmissions. 8 Meanwhile, for patients who have a hernia that can be readily repaired laparoscopically, RVHR is a substantially more expensive option for repair in the present study. However, the total charges for an open repair were higher than LVHR or RVHR.
In logistic regression, smoking and diabetes were independently associated with readmission. The authors hypothesize that obesity was not significant in the logistic regression analysis because obesity was defined as body mass index (BMI) of > 30 kg/m2. The relationship between obesity and postoperative complications is dose-dependent with patients who have a BMI of > 40 kg/m2 being at a marked increased risk for complications. 20 Patients with these comorbidities are known to be at a higher risk of wound complications, which are a predominant driver of readmissions. 21 In a recent study published in JAMA Open by Howard et al, the authors showed how remarkably uncommon appropriate optimization of hernia patients is despite the overwhelming evidence in the literature to support it. Indeed, the authors found that nearly one-third of patients had a high-risk characteristic (smoking, obesity, or unhealthy alcohol consumption). 12 There was also a significant variation between the hospitals included in the study. At some hospitals, nearly half of the patients had at least 1 high-risk characteristic at the time of surgery. A focus on preoperative optimization is one of the most important ways to improve patient outcomes. 22 Other factors that were predictive of readmission in the present study are not necessarily modifiable, such as emergent repair, but are important to consider when counseling patient on their operative risk. Urban-teaching hospital status, which has been previously linked to readmission, was again predictive of readmission, but patients at urban-teaching hospitals have been shown to be more comorbid than other hospital types. 23
This study is limited inherently by the use of a large-scale database. There are factors that are not identifiable with the NRD that are known to impact surgical outcomes. Size of the defect, recurrent status of the ventral hernia, mesh type, location for placement of the mesh, and Centers for Disease Control wound class are not contained in this data set. This may yield unfair comparisons between smaller hernias being fixed in one manner vs larger hernias in another. However, the primary goals of the study were achieved in the largest readmission analysis to date. Future directions of study should seek to evaluate readmissions over longer time periods as patients may be readmitted for postoperative complications far removed from surgery. Limitations within the database also include overall charges, which can be institution-specific. A cost analysis may be more meaningful when comparing open, laparoscopic, and robotic-assisted approaches. Additionally, as surgeons gain familiarity with the robotic platform, RVHR volume and the ability of to perform RVHR in more comorbid and more complex patients should increase. A continual evaluation of RVHR and its outcomes should be undertaken as its popularity in VHR increases, with controls for hernia characteristics leading to more conclusive comparisons.
Footnotes
Author’s Note
Gold Medal Oral Presentation at the annual Southeastern Surgical Congress, Atlanta, GA, August 2021
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
